Household versus Economist Forecasts of Inflation: A Reassessment: Note

Abstract

In a recent article in this Journal} Gramlich (1983) concluded that the mean inflation forecasts from the University of Michigan Survey Research Center (SRC) surveys of households were, in the years 1956-1980, more accurate and more rational than the mean inflation forecasts from Livingston's Philadelphia Inquirer surveys of professional economists. These paradoxical Elndings have proved robust with respect to improvements in testing procedures (Bryan and Gavin 1986a, 1 986b). The purpose of this note is not to cast doubt on the result themselves, but to urge caution in their interpretation. With respect to forecast accuracy, we show that while the average forecast from the SRC is indeed more accurate than the average forecast from the Livingston survey, the forecast of a typical household is much less accurate than that of a typical economist. The apparent superiority of the SRC survey is entirely due to its larger